1) ② crawl to the bright Web Part travel news

1 __author__='Minmin'2 #Coding:utf-83 ImportRe,urllib,sgmllib4 5 #get HTML based on the current URL6 defgethtml (URL):7page =urllib.urlopen (URL)8HTML =Page.read ()9 page.close ()Ten     returnHTML One  A #get the article content you want based on HTML - deffunc (str): -result= Re.findall (r"<p style=\ "text-indent:30px; margin:0px 3px 15px\ "> ([^<>]*) </p>", gethtml (URL), re. MorRe.findall (R"<p> ([^<>]*) </p>", gethtml (URL), re. M) the      #or Re.findall (r "<p style=\" Text-justify:distribute; Text-align:justify\ "align=\" justify\ "> (. *?) </p> ", gethtml (URL), re. M) -artical ="' -       forJinchResult: -          ifLen (j) <>0: +j = J.replace ("&nbsp;","") -j = J.replace ("<strong>","   ")#Remove the <strong>, and replace it with "" . +j = J.replace ("</strong>","   ")#Remove </STROGN> Replace "" Aartical = artical + j +'\ n' at      returnartical -  - #The HTML link is labeled "A", and the linked property is "href", which is to get all the TAG=A,ATTRS=HREF values in the HTML.  - classUrlpaser (sgmllib. Sgmlparser): -     defReset (blank): - Sgmllib. Sgmlparser.reset (blank) inBlank.urls = [] -  to     defstart_a (blank,attrs): +href = [v forKvinchAttrsifK = ='href'] -         ifhref: the blank.urls.extend (HREF) *  $Iparser =Urlpaser ()Panax NotoginsengSocket = Urllib.urlopen ("http://travel.gmw.cn/node_39034.htm")#Open this page -  the #fout = file (' Qq_art_urls.txt ', ' W ') #要把这个链接写到这个文件中 +Iparser.feed (Socket.read ())#analysis. A  theReg ='http://travel.gmw.cn/2015-.*' #this is used to match eligible links, using regular expression matching +Reg2='http://travel.gmw.cn/2014-.*' -Pattern =Re.compile (REG) $Patter =re.compile (REG2) $I=0 -Url2=[] -  forUrlinchIparser.urls:#links are available in URLs theURL ="http://travel.gmw.cn/"+URL -     ifpattern.match (URL):Wuyi         ifUrl not inchUrl2: the url2.append (URL) -             PrintURL Wuartical =func (URL) -             Printartical About             ifLen (artical) <>0: $i = i + 1 -f = open ("gmw/travel/"+str (i) +'. txt','A +') - f.write (artical) - f.close () A  +     ifpatter.match (URL): the         ifUrl not inchUrl2: - url2.append (URL) $             PrintURL the             Printartical the             ifLen (artical) <>0: thei = i + 1 thef = open ("gmw/travel/"+str (i) +'. txt','A +') - f.write (artical) inF.close ()

1) ② crawl to the bright Web Part travel news